BIOMECHANICAL RESPONSE REQUIREMENTS OF THE THOR … · THE THOR NHTSA ADVANCED FRONTAL DUMMY...

Report No: GESAC-05-03

BIOMECHANICAL RESPONSE REQUIREMENTS OFTHE THOR NHTSA ADVANCED FRONTAL DUMMY

(Revision 2005.1)

TRAUMA ASSESSMENT DEVICE DEVELOPMENT PROGRAM

Revised: March, 2005

Prepared under contract DTNH22-99-C-07007 for the National Transportation Biomechanics Research Center,Office of Human-Centered Research, National Highway Traffic Safety Administration, U.S. Department ofTransportation

General Engineering and Systems Analysis Company, Inc125 Orchard Drive

Boonsboro, MD 21713

i Biomechanical Response Requirements [2005.1]

BIOMECHANICAL RESPONSE REQUIREMENTS OF THE THORNHTSA ADVANCED FRONTAL DUMMY

1. INTRODUCTION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1

SECTION I. CURRENT BIOMECHANICAL REQUIREMENTS . . . . . . . . . . . . . . . . . . . 3

2. HEAD . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32.1 Head Drop Test . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32.2 Head Impact Test (Whole-Body) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4

3. NECK . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63.1 Neck Kinematic and Dynamic Requirements in Frontal Flexion . . . . . . . . . . . . . 63.2 Neck Kinematic and Dynamic Requirements in Lateral Flexion . . . . . . . . . . . . . 93.3 Quasi-static Response at Atlanto-Occipital Joint in Flexion/Extension . . . . . . . 12

4. THORAX . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 144.1 Sternal Impact . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 144.2 Lower Ribcage Oblique Impact . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16

5. ABDOMEN . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 195.1 Upper Abdomen: Steering Wheel Impact . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 195.2 Lower Abdomen: Rigid Rod Impact . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20

6. FEMUR . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22

7. FACE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 247.1 Rigid Bar Impact . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 247.2 Disk Impact . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25

8. LOWER LEG/ANKLE/FOOT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 278.1 Dynamic Heel Impact . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 278.2 Dynamic Dorsiflexion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 288.3 Quasi-static Inversion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 308.4 Quasi-static Eversion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 318.5 Quasi-static Dorsiflexion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 328.6 Quasi-static Plantarflexion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 348.7 Quasi-static Internal/External Rotation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35

SECTION II. BIOFIDELITY PERFORMANCE OF THOR . . . . . . . . . . . . . . . . . . 37

9. HEAD BIOFIDELITY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 379.1 Head Drop Test . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 379.2 Head Impact Test (Whole Body) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37

ii Biomechanical Response Requirements [2005.1]

10. NECK BIOFIDELITY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3810.1 Frontal Flexion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3810.2 Lateral Flexion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4110.3 A-O Joint Response . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44

11. THORAX BIOFIDELITY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4511.1 Sternal Impact . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4511.2 Lower Ribcage Oblique Impact . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46

12. ABDOMEN BIOFIDELITY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4812.1 Upper Abdomen: Steering Wheel Impact . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4812.2 Lower Abdomen: Rigid Rod Impact . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48

13. FEMUR BIOFIDELITY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50

14. FACE BIOFIDELITY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51

15. LOWER LEG/ANKLE/FOOT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5315.1 Dynamic Heel Impact . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5315.2 Dynamic Dorsiflexion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5415.3 Quasi-static Inversion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5515.4 Quasi-static Eversion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5515.5 Quasi-static Dorsiflexion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5615.6 Quasi-static Plantarflexion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5715.7 Quasi-static Internal/External Rotation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58

SECTION III. PROPOSED ADDITIONAL BIOMECHANICAL REQUIREMENTS. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59

16. NECK . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5916.1 Neck Kinematic and Dynamic Response in Extension . . . . . . . . . . . . . . . . . . . . 5916.2 Neck Axial Compression Dynamic Response . . . . . . . . . . . . . . . . . . . . . . . . . . . 6216.3 Neck Tensile Loading Requirement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6316.4 Neck Torsion Requirement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64

17. SPINE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66

18. THORAX . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6718.1 Thorax: Regional Coupling under Quasi-static Loading . . . . . . . . . . . . . . . . . . . 6718.2 Thorax: Regional Loading under Belt Impact . . . . . . . . . . . . . . . . . . . . . . . . . . . 69

19. ABDOMEN . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7019.1 Abdomen: Belt Impact . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7019.2 Abdomen: Airbag Impact . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71

iii Biomechanical Response Requirements [2005.1]

20. FACE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7320.1 Cylindrical Impactor Response . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73

21. LOWER LEG/ANKLE/FOOT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7521.1 Dynamic Inversion/Eversion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75

1 Biomechanical Response Requirements [2005.1]

BIOMECHANICAL RESPONSE REQUIREMENTS OF THE THORNHTSA ADVANCED FRONTAL DUMMY

1. INTRODUCTION

This report constitutes part of the final documentation for the NHTSA Advanced FrontalDummy known as THOR (Test device for Human Occupant Restraint). It describes thebiomechanical response requirements which have been defined to assess the biofidelity of theThor dummy. The report also provides results from tests performed to evaluate the level ofagreement with these requirements. A summary of the test procedures used for testing Thor withrespect to the requirements are given, but the exact details of the testing procedures will be foundin the Thor Certification Manual which is a part of the Thor documentation. In addition, thereport proposes a number of additional requirements, which will assess the biofidelity of Thor ina number of test environments which were not covered by the first set of requirements.

The report is arranged in three sections. In each section, the report is arranged according to theprincipal segments of the dummy. Section I contains the requirements that were defined duringthe initial design, fabrication, and testing of the dummy. These requirements have beenavailable for some time and the response of the Thor dummy has been tested against theserequirements during an extensive test program. For each segment, the sources of backgroundinformation are provided, followed by a brief description of the experimental procedure, andfinally, the target biomechanical response is presented.

Section II presents the results from testing performed on Thor to evaluate the biofidelity of thedummy with respect to the biomechanical requirements.

Section III contains requirements that have been more recently proposed and includes datagathered from more recent biomechanical research. The Thor dummy has either been tested onlyin a limited way, or not at all, against these newer requirements. Elements of these requirementsmay be adopted in fully qualifying the Thor ATD.

In Section I, biomechanical response requirements for the following segments are described.

1. Head - Based on cadaver drop tests by Hodgson and Thomas.2. Neck - Dynamic neck bending responses for frontal and lateral flexion

based on tests performed at Naval Biodynamics Laboratory andkinematic data developed by TNO, and moment-angle relationsdeveloped by Mertz, Patrick, and Chou.

3. Thorax - Upper ribcage requirement based on force-deflection datadeveloped by Neathery based on testing by Kroell ; lower ribcagerequirement based on force-deflection data developed by MCW.

5. Abdomen - Lower abdomen force-penetration data based on low severityimpacts carried out by Cavanaugh; upper abdomen force-penetration data based on impacts carried out by Nusholtz.

6. Femur - Based on knee impact tests performed and analyzed by Horsch and


Patrick.7. Face - Based on impact tests by Nyquist, Allsop, and Melvin and

analyzed by Melvin.8. Lower Leg - Dorsiflexion response from dynamic tests performed by Renault,

leg axial response from dynamic tests performed by MCW, and anumber of quasi-static responses in dorsiflexion, plantarflexion,inversion, and eversion, performed by UVa and Renault.

In Section II, as indicated above, the results from testing Thor according to the requirementsoutlined in Section I are presented.

In Section III, the proposed., additional response requirements are described.

1. Neck - Dynamic neck bending response in extension based on testsperformed by TNO and moment-angle relations developed byMertz and Patrick. Secondary requirements based on compression,tension, and torsion response.

2. Spine - No biomechanical requirements currently defined.3. Thorax - Quasi-static localized response data based on testing by

Cavanaugh, and belt impact response based on testing by INRETS.4. Abdomen - Force-deflection response for belt and airbag interaction with

abdomen based on testing by UMTRI.5. Face - Additional requirements based on more localized testing by

INRETS.6. Lower Leg - Dynamic inversion/eversion response based on testing by Renault.


SECTION I. CURRENT BIOMECHANICAL REQUIREMENTS

2. HEAD

References

Hodgson, V.R., Thomas, L.M. 1971. Comparison of Head Acceleration Injury Indices inCadaver Skull Fracture. Fifteenth Stapp Car Crash Conference. SAE 710854.Hubbard, R.P. McLeod, D.G. 1974. Definition and Development of a Crash Dummy Head. Eighteenth Stapp Car Crash Conference. SAE 741193.Melvin, J., Weber, K. 1985. Task B Final Report. DOT HS 807 042. NHTSA.Prasad, P., Melvin, J., Huelke, D., King, A., Nyquist, G. 1985. Head. Review of BiomechanicalImpact Response and Injury in the Automotive Environment. pp 1-43. Ed. J.

Two tests can be used to verify the response of the head to frontal impact. Both of these tests arebased on the head impact tests of Hodgson and Thomas at Wayne State University [1971]. Thefirst test is an isolated head drop test and the second is a whole body impact test with a movingimpactor which is meant to be equivalent to the head drop test. Thus either or both tests may beused to establish the biofidelity of the Thor head.

2.1 Head Drop Test

The forehead impact response is based on tests by Hodgson and Thomas [1971]. In these tests,the cadavers were strapped to a pallet that was free to rotate at the feet. The cadaver’s head andneck extended over the free end. The pallet was released at a fixed distance from a rigid impactsurface. The average peak acceleration was 250g and the average impact velocity was 2.71 m/s(107 in/sec) which is equivalent to a free fall height of 376 mm (14.8 in). A correction was madeto the impact velocities initially reported by Hodgson due to the rotation of the pallet, and uponfurther analysis, the impact velocities were revised upwards. Hubbard and McLeod [2] used thedata to generate a corridor for the head impact response by using an allowable variation of +/-10%, i.e. the range was 225-275g. The head drop test and the corresponding corridor became therequirement for establishing the biofidelity of the GM ATD 502 head [Hubbard, et al., 1974],which was later used as the head for the Hybrid III dummy.

Test Setup

1. Suspend the head such that it is oriented symmetrically about its mid-sagittal plane. 2. The head should be oriented about its lateral axis, so that the flattest part of the forehead

is approximately parallel to the impact surface. 3. The lowest point on the forehead should be 376 mm (14.8 in) above the impact plate. 4. Drop the head from the specified height, so that it falls freely onto a rigid, horizontal steel

plate.


Figure 1. Head drop test corridor (376 mm height).

Biomechanical Response

The corridor is described by an acceleration vs time response for the specified drop height, asshown in Figure 1.

2.2 Head Impact Test (Whole-Body)

Description

Melvin [1985] proposed an equivalent procedure which equated the impact energy in the headdrop tests, with the effective impact energy from an impactor test on the forehead of a completedummy. The procedure developed by Melvin has been selected to establish the biofidelity ofThor under head impact. This procedure has some advantages over the normal head drop test. The problems of orienting the head during a head drop test are avoided. In addition, thebiofidelity of impacts to the lateral aspect of the head is simpler to evaluate, as it requires asimple rotation of the dummy. The priority will be given to meeting the response for impactvelocity of 2.0 m/s.

Test Setup

1. The subject is set up in a sitting position, with its legs horizontal and the arms raised(may be taped lightly to a support bar).

2. Impactor mass is 23.4kg. The impact surface is rigid and flat, with a diameter of 15.2 cm.3. The impact speed is 2.0 m/s .4. The head of the dummy is placed, such that the axis of the impactor is aimed at a point on

the forehead on the midsagittal plane and 30 mm above the horizontal line marking thedivision with the face skin. The tilt of the dummy head/neck assembly is adjusted so thatthe impact area on the head is parallel to the face of the impactor.


Figure 2. Response of head for impact with rigid impactor[Melvin, 1985]


The corridors are described as a force vs duration time response for the two impact speeds andare shown in Figure 2.


3. NECK

References

Ewing, C.L., Thomas, D.J., Beeler, G.W., Patrick, L.M., and Gillis, D.B. 1968. DynamicResponse of the Head and Neck of the Living Human to -Gx Impact Acceleration. Twelfth StappCar Crash Conference. SAE Paper # 680792.Ewing, C.L., Thomas, D.J., Patrick, L.M., Beeler, G.W., and Smith, M.J. 1969. Living Human Dynamic Response to -Gx Impact Acceleration, Part II - Accelerations Measured on the Headand Neck. Thirteenth Stapp Car Crash Conference. Ewing, C.L., and Thomas, D.J. 1973. Torque Versus Angular Displacement Response for theHuman Head to -Gx Impact Acceleration. Seventeenth Stapp Car Crash Conference. SAE730976.Ewing, C.L., Thomas, D.J., Lustick, L., Becker, E., Willems, G., and Muzzy, W.H. 1975. TheEffect of the Initial Position of the Head and Neck on the Dynamic Response of the Human Headand Neck to -Gx Impact Acceleration. Nineteenth Stapp Car Crash Conference. SAE Paper #751157.Ewing, C.L., Thomas, D.J., Lustick, L., Muzzy, W.H., Willems, G., and Majewski, P.L. 1976.The Effect of Duration, Rate of Onset and Peak Sled Acceleration on the Dynamic Response ofthe Human Head and Neck. Twentieth Stapp Car Crash Conference. SAE Paper # 760800.Ewing, C.L., Thomas, D.J., Majewski, P.L., Black, R., and Lustick, L. 1977. Measurement ofHead, T1, and Pelvic Response to -Gx Impact Acceleration. Twenty-First Stapp Car CrashConference. SAE Paper # 770927.Klinich, K., Beebe, M., Pritz, H., and Haffner, M. 1995. Performance Criteria for a BiofidelicDummy Neck. National Highway Traffic Safety Administration, Vehicle Research and TestCenter.Mertz, H.J. and Patrick, L.M. 1971. Strength and Response of the Human Neck. Fifteenth StappCar Crash Conference. SAE Paper # 710855.Patrick, L.M., and Chou, C.C. 1976. Response of the Human Neck in Flexion, Extension andLateral Flexion. Vehicle Research Institute Report VRI 7.3.Thunissen, J. et al. 1995. Human Volunteer Head-Neck Response in Frontal Flexion: A NewAnalysis. Thirty-ninth Stapp Car Crash Conference. SAE Paper # 952721.Wismans, J., and Spenny, C.H. 1983. Performance Requirements for Mechanical Necks inLateral Flexion. Twenty-seventh Stapp Car Crash Conference. SAE Paper # 831613.Wismans, J., and Spenny, C.H. 1984. Head-Neck Response in Frontal Flexion. Twenty-eighthStapp Car Crash Conference. SAE Paper # 841666.Wismans, J. et al. 1987. Comparison of Human Volunteer and Cadaver Head-Neck Response inFrontal Flexion. Thirty-first Stapp Car Crash Conference. SAE Paper # 872194.

3.1 Neck Kinematic and Dynamic Requirements in Frontal Flexion

Description

The frontal flexion requirements arose from the extensive tests conducted by Ewing, et al. [1968,1969, 1973, 1975, 1976, 1977]. These data were later analyzed by Wismans and Spenny [1983,


1984, 1987], from which the final form of the requirements were developed. In the originalNBDL tests, two sled acceleration levels, at 15 G and 8 G, were used to develop responses forsagittal flexion. The response at the more severe 15 G level will be the primary responserequired to be met, since this is closest to the T1 accelerations that will be encountered withdecelerations generated at typical 48 kph or 56 kph impact speeds. T1 accelerations at theseseverities may be around 40 G.

The neck response is primarily prescribed by the time histories of the head rotation angle and thelongitudinal and vertical displacements of the head C.G. The latest revisions to the corridors forthe 15 G input pulse have been presented by Thunissen, et al. [1995]. These are similar to therequirements proposed by Klinich et al. [1995], which had formed the basis of an earlierrequirement, and is still used for defining the lateral flexion response.

In addition to primary responses described above, the resultant acceleration of the head C.G. asdefined by Thunissen is also used as kinematic requirement. For the dynamic requirement, themoment-angle response at the occipital condyle [O.C.] described originally by Mertz and Patrick[1971], is used.

Test Setup

The neck flexion tests were performed on the NBDL volunteers, according to the followingprocedure:1. The subject is seated upright on the sled.2. Subjects restrained by pair of shoulder belts, a lap belt, and an inverted-V pelvic strap

tied to the lap belt.3. The motion of the sled is adjusted to have a acceleration curve shown in Figure 2.4. Target markers and accelerometers were used to record the kinematic and dynamic

responses of the subject.

For testing the biofidelity of the Thor neck, the procedure is simplified by testing the head andneck assembly only, with the base of the neck being directly attached to a sled apparatus(optionally through a lower neck load cell) that can duplicate the acceleration pulse seen at theT1 location. There will be some discrepancy with the whole-body test, since the rotation at T1,seen in the volunteer testing, cannot be generated with this procedure. As indicated in Thunissen[1995], the error in neglecting T1 rotation is estimated to be small.

Biomechanical Requirement

The input acceleration pulse at T1 is shown in Figure 3 and the original sled pulse is shown inFigure 4 for reference. The responses for the 15 G frontal flexion test are shown in Figures 5 to8.


Figure 3. Input acceleration pulse for 15 G frontalflexion test of neck (equivalent to T1 acceleration ofvolunteers).

Figure 4. Input sled acceleration for volunteers for 15G frontal flexion test of neck.

Figure 5. Head rotation angle corridor for 15 Gfrontal flexion test.

Figure 6. Corridor for resultant acceleration of headC.G. for 15 G frontal flexion test.

Figure 7. Head C.G. X displacement corridor for 15 Gfrontal flexion test.

Figure 8. Head C.G. Z displacement corridor for 15frontal flexion test.


Figure 9. Moment-angle corridor for frontal flexion, measured at occipitalcondyle.

As indicated above, the dynamic response of the neck for frontal flexion is defined by thecorridors presented by Mertz and Patrick [1971]. The moment is measured about the occipitalcondyle, and the angle measures the angle of the head relative to the torso (with respect to theneutral head position). The frontal flexion corridor will implicitly contain the contribution fromany chin-chest contact which accounts for the large increase of the moments at the higher angles.Figure 9 shows the moment-angle corridor for flexion.

3.2 Neck Kinematic and Dynamic Requirements in Lateral Flexion

Description

The lateral flexion requirements are based on the same NBDL studies listed in section 2.2.1. Forthese requirements, the data presented by Klinich [1995] are used.

In the Klinich study, the corridors for the head and neck angles were provided. Anapproximation to the head C.G. displacement corridors were obtained in the following way:

The neck and head anthropometry provided in the paper by Klinich, was used to determine thevector location of the occipital condyle (O.C.) relative to T1 and the location of the head C.G.relative to the O.C. The location of the principal landmarks (in meters) used were:

T1: x = 0.; z = 0.O.C.: x = -.003; z = .119Head C.G.: x = .014; z = .058

These measurements were made with the head and neck aligned. The standard deviation in the


Figure 10. Input acceleration pulse for 7 G lateralflexion test of neck (equivalent to T1 acceleration ofvolunteers).

Figure 11. Input acceleration for 7G neck lateralflexion test

head and neck angles were then used to determine the relative deviation to be expected in thehorizontal and vertical displacements of the head C.G. at each time.

Test Setup

The setup for this test is given by:1. The subject is seated on the sled, facing laterally with respect to the sled moving

direction.2. Subjects restrained by pair of shoulder belts, a lap belt, and an inverted-V pelvic strap

tied to the lap belt. A 25 cm wide chest strap is used to minimize the load on the rightshoulder. In addition, a lightly padded wooden board is placed against the right shoulderto limit the upper torso motion.

3. The motion of the sled is adjusted to have a acceleration curve shown as Figure 9. 4. Use target marks and accelerometers to record the kinematic and dynamic responses of

the subject.

For testing the biofidelity of the Thor neck in lateral flexion, the procedure is simplified bytesting the head and neck assembly only, with the base of the neck being directly attached to asled apparatus (optionally through a lower neck load cell) that can duplicate the accelerationpulse seen at the T1 location. There will be some discrepancy with the whole-body test, sincethe rotation at T1, which may occur in the volunteer testing, cannot be generated with thisprocedure.


The input acceleration pulse at T1 is shown in Figure 10 and the original sled pulse is shown inFigure 11 for reference. The responses for the 15 G frontal flexion test are shown in Figures 12to 14.


Figure 12. Head angle rotation corridor for 7 G lateralflexion test.

Figure 13. Head CG Y displacement corridor for 7 Glateral flexion test.

Figure 14. Head CG Z displacement corridor for 7 Glateral flexion test.

The dynamic response of the neck for lateral flexion is defined by the corridors presented byPatrick and Chou [1971]. Figure 15 shows the moment-angle corridor for the neck moment atthe O.C. during lateral flexion.


Figure 15. Moment-angle corridor measured at occipital condyle for lateralflexion

3.3 Quasi-static Response at Atlanto-Occipital Joint in Flexion/Extension

Nightingale, R., et al. 2005. Flexion and Extension Structural Properties and Strength for MaleCervical Spine Segments. Journal of Biomechanics. (to be published)

Description

Testing was performed on unembalmed cervical spinal segments. Muscular tissues wereremoved, keeping all the ligamentous structures intact. The segments were sectioned into 4groups: O-C2, C3-C4, and C5-C7. Bending tests were performed in a pure-moment test frame. The flexibility curves were based on quasi-static loading at 0.5 Nm increments to 3.5 Nm inflexion and -3.5 Nm in extension. For the purposes of application to the Thor dummy, themoment-angle relationship for the O-C2 segment is used as a lower limit to the flexibility of theO.C. joint in the dummy (which approximately represents this portion of the neck). TheP.M.H.S. segment will not have any inter-segment musculature which will make the overallstiffness greater.

Nightingale, et al. fit a logarithmic function to estimate angle as a function of moment. For thepurposes of use in the Thor O.C. response, the function was extrapolated beyond the +/- 3.5 Nmmeasured values to make estimates for the O.C. response up to +/- 30° of flexion/extension. The measured standard deviation was also extrapolated to the portion of the curve outside the +/-3.5 Nm range.


Figure 16. Moment-angle response at A.O. joint (quasi-static).

Test Setup

The corresponding test setup for the dummy O.C. joint, is as follows:

1. Attach base of joint to a rigid frame.2. Apply moment to the top of the joint structure. In the dummy, the upper half of the joint

is situated in the head/neck mounting platform that forms the base of the head. A suitableinterface would be attached to the platform to allow application of the moment.


The following graph provides the quasi-static moment-angle response for the ligamentous neckexpected at the O.C. The graphs shows the average response along with +/- one standarddeviation. This should provide a lower limit to the O.C. joint response of the dummy.


4. THORAX

4.1 Sternal Impact

References

Kent, R., et al. 2004. The Role of Muscle Tensing on the Force-Deflection Response of theThorax and a Reassessment of Frontal Impact Thoracic Biofidelity Corridors. Proceedings ofthe 48th Stapp Car Crash Conference. Neathery, R. 1974. Analysis of Chest Impact Response Data and Scaled PerformanceRecommendations. Proceedings of the 18th Stapp Car Crash Conference.

Description

The principal response corridors for sternal impact are the traditional Kroell corridors for rigiddisk impacts at 4.3 m/s and 6.7 m/s at the upper ribcage. The normalized curves provided byNeathery [1974] defined the response requirements for this type of impacts for a 50th percentileU.S. male. Recently, it has been suggested [Kent, et al, 2004] that due to the relatively low rateof loading on the thorax, the effective stiffness of the thorax as estimated by Kroell and Neatheryshould be reevaluated. During the development of the original corridors by Neathery, a fixed “correction’ of 667 N (150 lb) was applied to the force in the force-deflection curves obtainedfrom the cadaver testing. This correction was meant to correspond to the stiffening that wouldoccur due to muscle tensing which would be expected to occur in live human occupants during acrash event. Kent, et al., have argued that a constant shift in force for all levels of chestdeflection is unrealistic, since it is unlikely that the thorax can remain tensed at the larger valuesof deflection. The authors have suggested, based on tests with untensed and tensed pigs, that amore appropriate force-deflection curve for the 4.3 m/s impact would be to remove the shift of667 N that was originally applied. This correction can remain in the 6.7 m/s test. NHTSA hasapproved the use of these modified corridors for the biofidelity requirements for the Thordummy, and these will be used provisionally as the new, revised requirements. NHTSA has alsospecified that the 4.3 m/s requirement will be the primary requirement for the dummy to meet,followed by the 6.7 m/s requirement.

Test Setup

The test configuration for performing the Kroell tests consists of:1. A rigid and flat impactor with a 152 mm diameter and mass of 23.4 kg. 2. The subject is set up in a sitting position, with no back support and its legs horizontal and

the arms raised. 3. The subject is positioned in front of impactor such that the center line of the impactor is

at the middle of ribs 4 and 5, near the mid-line of the sternum. 4. Two impact speeds are tested: one at 4.3 m/s and one at 6.7 m/s.5. The force-deflection characteristics are obtained from the test, including the unloading

portion of the force to estimate the amount of hysteresis.


Figure 17. Force-deflection response of thorax for central disk impact at 4.3m/s

Figure 18. Force-deflection response of thorax for central disk impact at 6.7m/s.


The sternal impact response is defined by the following two corridors representing the responseat the two impact speeds:


Figure 19. Force-deflections characteristics of padding used inoblique impacts of lower ribcage.

4.2 Lower Ribcage Oblique Impact

References

Yoganandan, N., Pintar, F., Kumaresan, S., Haffner, M., Kuppa, S. 1997. Impact biomechanicsof the human thorax-abdomen complex. International Journal of Crash, Vol 2, No. 2, pp 219-228

Description

This test is based on oblique impacts at the lower ribcage performed by Medical College ofWisconsin (MCW) [Yoganandan, 1997]. In these tests, the torso was initially rotated from rightto left by 15°, such that the impact occurred on the right antero-lateral thorax. Theinstrumentation in the MCW tests consisted of a load cell and uniaxial accelerometer attached tothe pendulum to measure the impact forces. The chest deflection was measured with a chestband which measured the external deformation of the thorax. The response characteristics of thelower ribcage are shown as a force-time corridor, a deflection-time corridor and a combinedforce-deflection corridor.

Test Setup

The test configuration for performing the MCW test consists of:1. The lower extremities are stretched horizontally and the upper extremities are extended

forward to allow positioning of the torso. The back of the torso was unsupported. 2. Impactor consisted of a 150 mm diameter rigid disk with a mass of 23.4 kg. The face of

the impactor was covered with a Rubatex R-451N padding of thickness 19 mm. Thepadding was rectangular in shape with each edge 150mm in length. The paddingcharacteristics are shown below.


Figure 20. Force-deflection response for lower ribcage/abdomen obliqueimpacts at 4.3 m/s.

3. Long underwear was placed on the subject to provide realistic seat interaction.4. The subject was seated facing the impactor on a thin teflon sheet. 5. The legs and arms were fully extended and the subject was seated on a low friction

surface.6. The impact velocity was 4.3 m/s.


There are three responses that are used to define the overall lower ribcage response to obliqueimpact. These are the force-deflection response, the force time history, and the deflection timehistory. These are shown in Figures 19-21. The primary requirement will be to meet the force-deflection response.

The above figure shows the average force-deflection curve along with the lower and uppercorridors for the force-deflection response (at ± 1 std dev). All three curves have the sameloading slope and approximately the same unloading slopes but different peak force anddeflection values.


Figure 21. Force-deflection response for lowerribcage/abdomen oblique impacts at 4.3 m/s.

Figure 22. Deflection-time response for lowerribcage/abdomen oblique impacts at 4.3 m/s.


5. ABDOMEN

5.1 Upper Abdomen: Steering Wheel Impact

References

Nusholtz, G., and Kaiker, P. 1994. Abdominal Response to Steering Wheel Loading. Proceedings of the 14th International Conference on Experimental Safety Vehicles.

Description

The response requirement for upper abdomen impact is derived from data developed by Nusholtz[1994] based on steering wheel impacts with engagement at region of L2. Six tests wereperformed with impact speeds of 3.9 m/s to 10.8 m/s with an average speed of 8.0 m/s. There isincreasing stiffness seen after a penetration of 80 mm, which probably arises from engagementof the steering wheel with the lower ribcage structure.

Test set-up

The test configuration for performing the upper abdomen impact test consists of:1. Impactor consists of 18 kg rigid steering wheel, rigidly mounted at an angle of 45°.2. The subject is set up in a sitting position, with no back support and its legs and the arms

raised (may be taped lightly to a support bar). 3. The lower thoracic spine is maintained in a vertical position and the leading edge of the

steering wheel should be at the level of upper abdomen.4. The impact speed should be 8.0 m/s.5. Impactor force should be obtained and the external penetration obtained using a linear

displacement transducer.


The force vs. penetration corridor is given in Figure 23.


Figure 23. Force-penetration response for upper abdomen torigid steering wheel impact at 8.0 m/s.

5.2 Lower Abdomen: Rigid Rod Impact

References

Cavanaugh, J., Nyquist, G., Goldberg, S., and King, A. 1986. Lower Abdominal Tolerance andResponse. Proceedings of the 30th Stapp Car Crash Conference.

Description

The response requirement for lower abdomen impact have been derived from the low severitytests performed by Cavanaugh [1986]. The requirement is in the form of a force vs externalpenetration corridor. The tests were conducted using a 25 mm diameter rigid bar of length 30cm and mass 32 kg, impacting perpendicularly the abdomen of cadavers at vertical location ofapproximately L3. Five tests were performed in the speed range of 4.9 to 7.2 m/s with anaverage impact speed of 6.1 m/s.

Test Setup

The test configuration for performing the abdomen impact test consists of:1. Impactor is a 25 mm diameter, 30 cm long rigid cylindrical rod with mass of 32 kg2. The dummy is set up in a sitting position, with no back support and its legs horizontal

and the arms raised (may be taped lightly to a support bar). 3. The lower thoracic spine is maintained in a vertical position and the center line of the

impactor was approximately at the level of L2.4. The average impact speed was 6.1 m/s (for the low velocity impacts).


Figure 24. Force-penetration response for lower abdomen impactwith rigid rod at 6.1 m/s.


The force vs. penetration corridor is given in Figure 24.


6. FEMUR

References

Horsch, J., and Patrick, L. 1976. Cadaver and Dummy Knee Impact Response. Proceedings ofthe 20th Stapp Car Crash Conference. SAE Paper # 760799.

Description

This requirement defines the response of the femur to axial impacts at the knee. The responsesare based on the tests and analysis of Horsch and Patrick [1976]. They conducted knee impactsalong the femoral axis of unembalmed male cadavers with a rigid pendulum impactor. Only thewhole body impact tests are used for defining the requirement.

The primary requirements define the peak knee impact forces generated when impactors ofdifferent masses with different impact speeds are allowed to impact the knee. Horsch andPatrick expressed the combined effect of both mass and speed by using a single variable whichwas proportional to the initial energy of the impactor system. The impactors were a series of flatfaced, rigid pendulum impactors. Their face diameters varied between 38 mm to 152 mm andtheir weights varied between .24 kg to 25 kg. The tests using the 25 kg impactors are notincluded in the requirement definition since these were far larger than any of the other impactorsizes used. The pendulums were dropped from heights of .1 m to .8 m. corresponding to impactspeeds of 1.4 m/s to 4.0 m/s.

Test Setup

The test configuration for performing the femur impact test consists of:1. Impactor is a 75 mm diameter, rigid, flat faced with mass of 5 kg2. The subject was seated on a flat seat with no back support with the lower leg hanging

over the seat edge and the feet resting lightly on the floor. The lower leg made an anglebetween 90° - 105° relative to the femur.

3. the left or right knee was placed such that the center of the knee was in line with the axisof the impactor and this axis was approximately in line with the femur axis.

4. Tests are conducted at two impact speeds: namely 2.0 m/s and 3.0 m/s5. Measurements include the impact force on the knee as measured by the impactor and the

acceleration along the axis of the femur as measured by an accelerometer placedproximal to the knee.

Knee Impact Calculation Procedure

The single independent variable defined by Horsch and Patrick which is a function of dropheight and reduced mass was based on a two mass model of the impactor and the subject knee. The effective subject mass is determined by them by fitting a straight line to the force vsacceleration curve generated from the knee impact. Thus:


Figure 25. Knee impact response in whole body configuration forvarying impactor mass and velocity.

Fknee = ms.aleg

where: Fknee = impact force measured at kneealeg = acceleration measured on legms = effective subject mass

With this value of effective mass, a new independent variable is defined:

Ee = h. mp ms/(mp + ms)

where: h = drop height of pendulum = v2/2g (v = impact speed)mp = mass of pendulum

This variable will be termed the effective impact energy. The term mp ms/(mp + ms) is thereduced mass of the two mass system. For the rigid mount case, the subject mass is very largeand the reduced mass is effectively equal to mp. For the whole body case, the acceleration timehistory of the proximal femur is used to determine the effective femur mass which should beused in the calculation of the effective impact energy described above. During the period whenthe knee is being loaded by the impactor, the impact is approximated by the two mass system(impactor and femur). The effective femur mass is determined from a regression fit to theequation: Ffem = mfemafem


The knee impact response for the whole body configuration is given by Figure 23.


7. FACE

7.1 Rigid Bar Impact

References

Allsop, D., Warner, C., Wille, M., Schneider, D., and Nahum, A. 1988. Facial Impact Response -A Comparison of the Hybrid III Dummy and Human Cadaver. Proceedings of the 32nd StappCar Crash Conference. Melvin, J., and Shee, T. 1989. Facial Injury Assessment Techniques. Proceedings of the 12thInternational Conference on Experimental Safety Vehicles.Nyquist, G., Cavanaugh, J., Goldberg, S., and King, A. 1986. Facial Impact Tolerance andResponse. Proceedings of the 30th Stapp Car Crash Conference.

Description

The facial impact response here is based on rod impacts performed by Nyquist, et al. [1986],Allsop, et al. [1988], and Melvin and Shee [1989]. The response requirements are in the form offorce vs time and force vs deflection curves for the rod impact. The response requirements havebeen summarized by Melvin [1989].

The requirement defines the approximate force-deflection behavior of the face upon impact witha rigid rod. The Nyquist tests used a 25 mm diameter rod, with an attached mass of 32 kg or 64kg which impacted the faces of unembalmed cadavers across the nose and zygoma. The averagecontact velocity for the tests was 3.6 m/s (with a range of 2.8 to 4.8 m/s). The Allsop tests useda 20 mm diameter bar composed of semi-circular disks. The impact mass was 14.5 kg and wasdropped from heights of 305 to 610 mm (corresponding to impact speeds of 2.45 to 3.46 m/s). Impacts were conducted on the zygoma and the maxilla.

The critical force part of the above force-deflection curve is the portion between about 12 mmand 25 mm. The region before this, as indicated by Melvin, indicates the response of the facialskin and other soft tissue and can be highly variable. The steeper section of the curvecorresponds to the loading of the facial skeletal structure.

Test set-up

For the rod impact test to the face, the configuration is defined as:1. The torso of the subject is maintained straight and the head vertical, with the legs kept

horizontal and arms raised.2. The rod and impactor assembly will have a total mass of 32 kg.3. The impact speed is set to 3.6 m/s (average of the range 2.4 - 4.8 m/s in cadaver testing).4. The rod is configured to impact along the mid-line of the left and right maxilla plates on

face.


Figure 26. Force-time response for facial impact with rigid rod tozygomatic region.


The force vs time response for impact by a rigid bar across the zygomatic region with impactspeeds between 2.8 m/s and 4.8 m/s is given by the following figure. (The figure below wasbased solely on the data presented by Nyquist, since Allsop did not present force vs time curves.)

7.2 Disk Impact

References

Allsop, D., Warner, C., Wille, M., Schneider, D., and Nahum, A. 1988. Facial Impact Response -A Comparison of the Hybrid III Dummy and Human Cadaver. Proceedings of the 32nd StappCar Crash Conference. Melvin, J., and Shee, T. 1989. Facial Injury Assessment Techniques. Proceedings of the 12thInternational Conference on Experimental Safety Vehicles.Nyquist, G., Cavanaugh, J., Goldberg, S., and King, A. 1986. Facial Impact Tolerance andResponse. Proceedings of the 30th Stapp Car Crash Conference.

Description

The facial impact response here is based on disk impacts performed by Nyquist, et al. [1986],Allsop, et al. [1988], and Melvin and Shee [1989]. The response requirement is in the form offorce vs time curve for the disk impacts. The response requirements have been summarized byMelvin [1989].


Figure 27. Force-time response of facial impact by disk.

Test set-up

For the disk impact, the configuration is defined as:1. The torso of the subject is maintained straight and the head vertical, with the legs kept

horizontal and arms raised.2. The disk and impactor assembly will have a total mass of 13 kg.3. The impact speed is set to 6.7 m/s.4. The center of the disk is configured to impact at the mid-point of the line joining the two

maxilla plates on the face.


The force vs time response for distributed loading due to a 152 mm diameter disk is given byFigure 25.


8. LOWER LEG/ANKLE/FOOT

8.1 Dynamic Heel Impact

References

Kuppa, S., Klopp, G., Crandall, J., et al. 1998. Axial Impact Characteristics of Dummy andCadaver Lower Limbs. 16th International Technical Conference on the Enhanced Safety ofVehicles. Vol. 2, pp 1608-1617.

Description

The response of the lower leg to impact on the plantar surface of the foot is based on the testsconducted at the Medical College of Wisconsin [Kuppa, 1998]. In these tests the rotation of thefoot was minimized. The lower limb was amputated at the knee and attached to a mini-sled withthe total mass of 16.8 kg. A pendulum of mass 23.5 kg was used to the impact the plantarsurface of the foot. The mini-sled was free to move on the rails after impact. Tests wereconducted with impact speeds from 2.2 m/s to 5.6 m/s.

Test Setup

For the axial impact along the lower leg, the configuration is defined as:• The lower leg specimen (with foot) was rigidly attached at the proximal tibia to a

minisled of total mass 16.8 kg. The sled was allowed to move freely on horizontal rails.• The plantar surface of foot was impacted with a 23.5 kg impactor with impact speeds in

the range of 2.2 m/s to 5.6 m/s.• Impact force at the heel and force at the proximal tibia were measured.


The biomechanical response is defined by the peak impact force seen for each impact velocity. This is plotted in the following graph.


Figure 28. Peak force response for axial impacts through heel with varyingimpact energies.

8.2 Dynamic Dorsiflexion

References

Crandall, J., Portier, L., Petit, P., et al. 1996. Biomechanical Response and Physical Propertiesof the Leg, Foot, and Ankle. Proc. 40th Stapp Car Crash Conf., SAE Paper No. 962424, pp 173-192.

Description

The dynamic dorsiflexion response is given by the moment vs. angle characteristics at the ankle,when the ball of the foot is impacted . The response was obtained from cadaver tests conductedat Renault [Crandall, 1996; Portier, 1997]. Only the data during loading were considered, since,in addition to dorsiflexion, the foot would tend to invert or evert during the unloading phase. The data are derived from the right and left legs of only one subject.

Test Setup

1. Foot set at about 90 deg relative to tibia.2. Impact ball of foot at 5 m/s.3. Measure forces and moments at distal tibia and compute moment at ankle, moment

through the ankle cross-section, and the Achilles tension force.


Figure 29. Moment-angle response for total ankle section indynamic dorsiflexion

Figure 30. Moment-angle response at ankle only in dynamicdorsiflexion.


The moment-angle response corridor estimated from the cadaver tests is shown in the followinggraph.


8.3 Quasi-static Inversion

References


Petit, P., et al. 1996. Quasi-static Characterization of the Human Foot-Ankle Joints in aSimulated Tensed State and Updated Accidentological Data. Proc. of the 1996 IRCOBIConference.

Description

Quasi-static moment vs angle response when the foot is rotated in inversion was obtained fromtests conducted by Renault [Petit, 1996]. Similar tests were conducted with human volunteers byUniversity of Virginia [Crandall, 1996]. In the Renault tests, the cadaver legs were amputated15 cm above the malleoli and the proximal ends of the tibia and fibula were fixed in bonecement. Rotation of the foot was accomplished by applying a torque to the calcaneous through ashaft attached to it. The rotations in the UVa tests were performed using a special six degree-of-freedom fixture that allows rigid rotations of the foot relative to the tibia along any specifiedaxis. The proposed requirement is based, currently, on the Renault data, since the UVa volunteershowed fairly large maximum inversion angles (greater than 50 degrees), which may not berealistic.

Test Setup

1. Set foot at 90° with respect to the tibia.2. With the tibia held fixed, apply a moment to rotate the foot in inversion. The moment

can be applied using an external force acting at the end of the foot.3. Measure moment at ankle (using lower tibia load cell or external load cell) and angle of

rotation.


The moment-angle response for quasi-static inversion is given by the following graph. This isderived from the data from the Renault cadaveric tests.


Figure 31. Moment-angle response for quasi-static inversion.

8.4 Quasi-static Eversion

References


Petit, P., et al. 1996. Quasi-static Characterization of the Human Foot-Ankle Joints in aSimulated Tensed State and Updated Accidentological Data. Proc. of the 1996 IRCOBIConference.

Description

Quasi-static moment vs. angle responses in eversion were obtained from tests conducted byRenault [Petit, 1996] using cadaveric subjects and by UVa using volunteers [Crandall, 1996]. The procedures are described in the previous section on the quasi-static inversion requirement. The results from both labs are similar and a single moment-angle curve can be derived from bothsets of data.

Test Setup

1. Set foot at 90° with respect to the tibia.2. With the tibia held fixed, apply a moment to rotate the foot in eversion. The moment can

be applied using an external force acting at the end of the foot.


Figure 32. Moment-angle response for quasi-static eversion.

3. Measure moment at ankle (using lower tibia load cell or external load cell) and angle ofrotation.


The moment-angle response in quasi-static eversion is given by the following graph.

8.5 Quasi-static Dorsiflexion

References


Description

University of Virginia performed cadaver and volunteer tests using the test fixture used in theinversion and eversion testing. [Crandall, 1996]. The volunteer tests were conducted with thesubjects in a relaxed state. The lower leg was restrained using nylon straps, while the foot wasplaced in an athletic shoe, which was restrained to a foot plate using straps. Tests wereconducted with knee flexion at 0° and 90° .

The lower legs of the cadavers were amputated at the mid-diaphysis of the femur and all themusculature and joints influencing the ankle joints were retained. Knee flexion was maintained


Figure 33. Moment-angle response for ankle withAchilles, for quasi-static dorsiflexion.

Figure 34. Moment-angle response for ankle only, forquasi-static dorsiflexion.

at either 0° or 90°. Quasi-static dorsiflexion tests were also performed by Renault. In thesetests, the cadaver legs were amputated 15 cm above the malleoli and the proximal ends of thetibia and fibula were fixed in bone cement. Muscle tension of 900 N was simulated by applyinga constant tension in the Achilles tendon.

The proposed requirement is based on the volunteer response with the knee flexed at 90°, sinceprobably best imitates the natural state of the ankle in a driving environment.

Test Setup

1. Set foot at 90° with respect to the tibia.2. With the tibia held fixed, apply a moment to rotate the foot in dorsiflexion. The moment

can be applied using an external force acting at the end of the foot.3. Measure moment at ankle (using lower tibia load cell or external load cell) and angle of

rotation.


The moment vs. angle responses under quasi-static dorsiflexion are given by the followinggraphs. The first graph shows the response for the complete ankle section, including the effectof the Achilles, while the second graph shows the response only of the ankle joint.


8.6 Quasi-static Plantarflexion

References

Paranteau, C. et al. 1996. A New Method to Determine the Biomechanical Properties of Humanand Dummy Joints. Proc. of the 1995 IRCOBI Conference.

Description

Paranteau [1996] determined quasi-static plantarflexion response using cadaver lower legs whichhad been excised at the distal tibia-fibula. The moment-angle response correspond to the effectof the ankle only, since there is no passive response from the musculature of the anterior cruralcompartment which had been severed. UVa conducted plantarflexion tests with volunteers in arelaxed state. These tests show much higher responses than the Paranteau tests. The likelyreason is the effect of the passive and active muscles in the volunteer response. The data alsoindicated that there was no or very low moment for the first 25°. The proposed requirement willuse the Paranteau data, with an initial flat response up to 25°.

Test Setup

1. Set foot at 90° with respect to the tibia.2. With the tibia held fixed, apply a moment to rotate the foot in plantarflexion. The

moment can be applied using an external force acting at the end of the foot.3. Measure moment at ankle (using lower tibia load cell or external load cell) and angle of

rotation.


The moment vs. angle response in quasi-static plantarflexion is given by the following graph.


Figure 35. Moment vs. angle response for quasi-static plantarflexion.

8.7 Quasi-static Internal/External Rotation

References

Siegler, S., et al. 1988. The Three-Dimensional Kinematics and Flexibility Characteristics of theHuman Ankle and Subtalar Joints - Part I: Kinematics. Transactions of the ASME, Vol 110.

Description

The quasi-static response in internal and external rotation is derived from tests performed bySiegler [1988]. This response is not a requirement in the ankle design of Thor, but used as aguide to provide a smooth response when the foot is rotated about the Z-axis (instead of a zeromoment response up to the point of contact and a sharp and high response beyond that).

Test Setup

1. Set foot at 90° with respect to the tibia.2. With the tibia held fixed, apply a moment to rotate the foot in internal and external

rotation.3. Measure moment at ankle (using external load cell) and angle of rotation.


The moment vs. angle response is provided by the following graph:


Figure 36. Moment vs. angle response in quasi-static internal and externalrotation.


SECTION II. BIOFIDELITY PERFORMANCE OF THOR

9. HEAD BIOFIDELITY

The Thor forehead was impacted according to both the setups described in Section 2.

9.1 Head Drop Test

The response of the Thor head in the head drop test is shown below.

9.2 Head Impact Test (Whole Body)

Both the time to reach peak impact force, and the magnitude of the peak force is within thecorridors established for the impact. The peak force is at the lower end of the corridor, perhapsreflecting the fact that part of the impacting surface of the 150mm disk will engage the softerface skin.


Figure 38. Response of Thor to forehead impact by 23.4 kg impactor at 2m/s.

10. NECK BIOFIDELITY

The neck biofidelity was evaluated in testing at several laboratories using a head and neckassembly attached to a HyGe sled. As indicated in Section 3, applying an accerelation to thebase of the neck equivalent to the measured T1 acceleration in human volunteer tests, is areasonable simulation of the whole body tests.

10.1 Frontal Flexion

The following are results obtained from testing at TNO equivalent to the 15 G frontal flexionsled tests carried out on NBDL volunteers.


Figure 39. Head rotation angle time history for neck frontal flexion test.

Figure 40. Head CG X displacement time history for neck frontal flexiontest.


Figure 41. Head CG Z displacement time history for neck frontal flexiontest.

Figure 42. Head CG resultant acceleration time history for neck frontalflexion.


Figure 43. Moment vs. angle response at O.C. for neck frontal flexion test.

It is seen that the Thor neck meets the required corridors in frontal flexion though generally atthe higher end of the head C.G. displacement corridor. This is also borne out by the moment-angle response seen in Figure 38, where the Thor neck lies just at the bottom of the requiredcorridor.

10.2 Lateral Flexion

The following are results obtained from tests conducted at TNO, equivalent to the 7G NBDLlateral flexion tests.


Figure 44. Head rotation angle time history for lateral flexion test.

Figure 45. Head CG Y displacement time history for lateral flexion test.


Figure 46. Head CG Z displacement time history for lateral flexion test.

Figure 47. Moment-angle response at O.C. for lateral flexion test.

It is seen that the Thor neck lies slightly outside the head angle and the head C.G. Ydisplacement corridors. The moment-angle response indicates that the neck is within thecorridor up to 45°.


Figure 48. Moment-angle response of O.C. joint in comparison to scaledPMHS data.

10.3 A-O Joint Response

The quasi-static A-O joint response of Thor is given below.

The THOR response is stiffer than the PMHS data, though it is in relatively closer agreement inextension. Since the THOR neck is shown to be generally biofidelic in flexion, no furthermodification was made to lower the OC stiffness in flexion.


Figure 49. Response of Thor to sternal impact during Kroell testing at 4.3m./s impact speed (modified).

11. THORAX BIOFIDELITY

11.1 Sternal Impact

Extensive series of Kroell-type tests have been performed to evaluate the response of Thor forcentral, sternal impact. The responses for the 4.3 m/s impact for four different Thor dummies areshown in the following graphs. The deflections of the upper ribcage are measured by theaverage of the upper left and right Crux measurement system in Thor. It should be noted that thecorridor is based on the modified Kroell corridor described in Section 4.1

The peak force, maximum deflection, hysteresis, and general shape of the force-deflection curveare reproducible . There is some variation in the force-deflection response in the intermediaterange of the curves.

The corresponding responses for Kroell tests at impact speed of 6.7 m/s are shown below.


Figure 46. Response of Thor to sternal impact during Kroell testing at 6.7/simpact speed.

For the 6.7 m/s tests, the final peak force is seen to be greater than that in the corridor, thoughthe curves are reproducible. It was indicated in Section 4.1, that the priority for sternal impactbiofidelity was to ensure the appropriate response of the 4.3 m/s impact (modified).

11.2 Lower Ribcage Oblique Impact

The response of Thor to oblique impact at the lower ribcage (at impact angle of 15° ), asdescribed in Section 4.2 (MCW test), is shown in the following graph. The penetration ismeasured using a linear potentiometer attached to the impactor. The external deflectionmeasurement is used in the Thor tests to compare with the external deflection measured on thecadavers using the chestband in the MCW tests. In the normal certification of the Thor lowerribcage only the internal deflection measurement from the Crux will be used.


Figure 51. Response of Thor to lower ribcage oblique impact at 4.3 m./simpact speed.

The maximum penetration seen in Thor is near the maximum value defined by the corridor. Thepeak force is higher than seen in the cadaver tests (3500 N in Thor vs. 2800 N for cadavers).


Figure 52. Response of Thor to steering wheel impact at 8.0 m/s to upperabdomen..

12. ABDOMEN BIOFIDELITY

12.1 Upper Abdomen: Steering Wheel Impact

The response of Thor to a steering wheel impact to the upper abdomen is shown below.

The repeatability of the response is good and there is general agreement up to 80 - 90 mm ofpenetration.. Beyond this level of penetration, the impact force increases rapidly, well outsidethe range seen in the cadaver tests. Thus the dummy is biofidelic for penetrations less than 90mm..

12.2 Lower Abdomen: Rigid Rod Impact

The response of Thor to impact by a 25 cm diameter rigid rod as described in Section 5.2 isshown in the following graph. The penetration is measured by the displacement of the impactor.


Figure 53. Response of Thor to rigid rod impact at 6.1 m/s to lowerabdomen..

As in the case of the upper abdomen, the force-deflection characteristics of the Thor abodmenare biofidelic up to 75 - 80 mm of penetration.


Figure 54. Response of Thor to axial impacts at the knee atdifferent impact velocities (right knee).

Figure 55. Response of Thor to axial impact at the knee at differentimpact velocities (left knee).

13. FEMUR BIOFIDELITY

The biofidelity of the Thor knee and femur assembly is determined using the knee impact test ofHorsch and Patrick, as described in Section 6. The response is provided as peak impact forcesmeasured for different impact velocities. This is plotted in the graphs below. The plots show theresponses from four different Thor dummies, and for both the right and left knees.


Figure 56. Response of Thor face to rigid rod impact at 3.6 m/s.

14. FACE BIOFIDELITY

There has been limited testing of the Thor face under the test conditions proposed by Melvin,and described in Section 7.1 and 7.2. The response of the face to rod impacts is shown in thefollowing graph.

The responses indicate that the peak forces are well within the human corridors, but that time toreach peak force is delayed by about 2.5 msec.

For the disk impact, the response of the Thor face is shown by the following graph.


Figure 57. Response of Thor face to 150 mm disk impact at 6.7 m/s.

In this case, the peak impact force is some above the corridor, while the timing is within.


Figure 58. Response of Thor-Lx to heel impacts at velocities in range of 2.0m/s to 4.0 m/s (with impactor mass of 8.6 kg).


The biofidelity of the Thor lower leg, ankle, and foot assemblies have been extensivelyevaluated. The responses are described below.

15.1 Dynamic Heel Impact

The response of theThor-Lx (lower extremity of Thor) to an axially directed impact to the heelis shown in the following figure. The test condition is not exactly the same as that described inSection 8.1. In these tests, the proximal end of the tibia was attached to a rigid platform, ratherthan a mini-sled. Thus the effective moving mass was simply the mass of the impactor.

UVa performed a series of pendulum impact tests to the plantar surface of the foot, whichduplicated the setup used in their cadaver studies. Though, this test was not used as arequirement to establish biofidelity, it was used to verify that the lower leg /ankle /foot assemblywas performing properly in an environment where the tibia would be acted on by a dynamicimpact force. The following graph shows the comparison of the response of the lower tibia loadcell ( Fz) with that measured in a cadaver.


Figure 59. Comparison of Thor-Lx Fz response in tibia under dynamicaxial loading with

Figure 60. Moment-angle response in dynamic dorsiflexion.

15.2 Dynamic Dorsiflexion

The dynamic dorsiflexion response for the Thor-Lx is shown in the following graph. Themoments calculated are at the ankle only, since the instrumentation for measuring Achillestension was not available. The test setup used a simple impact pendulum with a mass of 5 kg,with a cylindrical impact face, which was allowed to impact the ball of the foot.


Figure 61. Moment-angle response in quasi-static inversion.

15.3 Quasi-static Inversion

The response in quasi-static inversion was obtained using a special fixture which is described inthe Thor Certification Manual. The fixture allows the rotation of the foot relative to the tibiaalong each of the ankle axis.

15.4 Quasi-static Eversion

The response of Thor in quasi-static eversion is shown in the following graph. The fixture usedin inversion, is also used here.


Figure 62. Moment-angle response in quasi-static eversion.

15.5 Quasi-static Dorsiflexion

Quasi-static dorsiflexion response is shown in the following graph. This response is not arequirement, since the ankle was designed to show appropriate response under dynamic loadingas shown in Figure 55. The response shown is for the whole ankle section (i.e. including theinfluence of the Achilles). It is seen, that under quasi-static loading conditions, the Thor anklehas a stiffer response than human volunteers.


Figure 63. Moment-angle response at ankle section under quasi-staticdorsiflexion.

Figure 64. Moment-angle response of ankle in quasi-static plantarflexion.

15.6 Quasi-static Plantarflexion

The response of the Thor ankle under quasi-static plantarflexion is shown in the following graph.


Figure 65. Moment-angle response in quasi-static internal/externalrotation.

15.7 Quasi-static Internal/External Rotation

The response of the Thor ankle assembly in quasi-static internal and external rotations is shownin the following graph.

The internal/external joint in the Thor-Lx was designed to provide approximate biofidelitywithin the range of ±15°. Beyond this point, the ankle is meant to stiffen up in a graduallyincreasing fashion.


SECTION III. PROPOSED ADDITIONAL BIOMECHANICAL REQUIREMENTS

16. NECK

16.1 Neck Kinematic and Dynamic Response in Extension

References

Ono, K. et al. 1999. Relationship between Localized Spine Deformation and CervicalVertfebral Motions for Low Speed Rear Impacts Using Human Volunteers. 1999 IRCOBIConference. pp 149-164.

Description

The requirements proposed here are based on the study carried out by Ono et al. [1999]. Theystudied the response of seven human volunteers under low speed rear-end impact. Subjects wereplaced on a seat mounted on a small sled system moving on an inclined rail which allowed theseat to slide freely backwards. The system was capable of sliding freely on a 4 meter long railinclined at an angle of 10°. At the end of travel, it collided with a damper. Tests wereconducted at impact speeds of 4, 6, and 8 km/h. A oil shock damper was installed to simulatethe accelerations generated during actual real impacts. At the 8 km/h impact, the peak sledacceleration was about 4 G. Both a rigid seat made of wood, and a standard, commercial seatwere used. A grip handle was installed in front of the seat to reproduce the extent of eachsubject’s muscular tension.

No headrest was used. The change of the spine configuration was measured by a new spinedeformation sensor with 33 paired strain gauges and placed against the spine. The interface loaddistribution was measured using a Tekscan system running at 100 frame/sec, placed between theseat and the subject. Cineradiography at 90 fps was also used . Measurements included two setsof biaxial accelerations on the head. Markers were placed on the head, neck, and torso to allowfor digitization. Motions were measured with high speed camera. The digitized informationincluded: rotation angles of the head, neck, and the upper torso. Spine extension was definedand analyzed as the change in length of the line connecting the neck-torso joint and the iliaccrest.

For the biofidelity requirement for Thor, the response at 8 kph impact speed and rigid seat isused as it offers the highest level of severity and the rigid seat ensures reproducibility of the testcondition.

Test Setup

1. Subject seated on a rigid seat sliding freely on a rail system inclined at 10° withoutheadrest.

2. Impact speed of 8 km/h.3. Measure accelerations to allow computation of the acceleration of the head C.G. and


Figure 66. Corridor for T1 acceleration for extensiontests.

Figure 67. Corridor for sled acceleration for extensiontests.

Figure 68. Corridor of head angle w.r.t T1 for sagittalextension

Figure 69. Corridor for Head CG X displacementw.r.t. T1 for extension.

angular acceleration of head. Determine motion of markers to allow determination ofhead angle time history.

For the dummy, a simpler test condition can be defined using only the head-neck system. In thiscase, the T1 acceleration measured with volunteers, is used as input to the base of the neck.


The acceleration at T1, which could be input to a head/neck assembly is shown in Figure 64. The original sled acceleration is shown in Figure 65. The neck kinematic responses in extensionare defined by the Figures 66-69, including head acceleration in the X direction.


Figure 70. Corridor for Head CG Z displacement w.r.tT1 for extension

Figure 71. Head X acceleration corridor for sagittalextension

Figure 72. Moment-angle corridor measured at occipital condyle forextension.

The dynamic response of the neck for sagittal extension is defined by the corridors presented byMertz and Patrick [1971].

Figure 72 shows the moment-angle corridor of neck during sagittal extension.


16.2 Neck Axial Compression Dynamic Response

References

Pintar, F.A., Yoganandan, N., Voo, J., Cusick, J.F., Maiman, D.J. and Sances, A., Jr., 1995.Dynamic Characteristics of the Human Cervical Spine, 39th Stapp Car Crash Conference, PaperNo. 952722.Pintar, F.A., Sances, A., Jr., Yoganandan, N., Reinartz, J., Mainman, D.J., Suh, J.K., Unger, G.,Cusick, J.F. and Larson, S.J., 1990. Biodynamics of the Total Human Cadaveric Cervical Spine,34th Stapp Car Crash Conference, Paper No.902309Nightingale, R.W., McElhaney, J.H., Camacho, D.L., Kleinberger, M., Winkelstein, B.A. andMyers B.S., 1997. The Dynamic Response of the Cervical Spine: Buckling, Ending Conditions,and Tolerance in Compressive Impacts, 41st Stapp Car Crash Conference, Paper No. 973344.

Description

This requirement is used as a secondary requirement for defining the response of the neck. Therequirement is based on the study by Pintar et al. [1995]. The tests were performed using acrown impact to the head at speeds ranging from 2.5 m/s to 8 m/s, using 20 unembalmed cadaverhead-neck complexes. The failure loads from this study ranged from 744N to 6431N (the meanfailure force is 3326N and the standard deviation is 1423N). The deformation at failure was 18±3mm. For this test configuration, a force-deformation corridor was derived and given in Figure70.

Nightingale et al. [1997] also performed a series of neck axial compression tests usingunembalmed human head and neck specimens with a drop track system. However, they did notdefine a corridor for the tests, but gave some numerical guidelines. Test Setup

1. The head-neck complex is placed vertically, with the inferior of the neck fixed.2. The neck is kept straight (pulley and deadweight can be used for this purpose).3. The diameter of the impactor surface is within 100mm-150mm, and the surface is padded

with a 20mm thick ensolite padding.4. The impact is delivered to the crown of the head.5. The impactor velocity is between 2.5m/s-8m/s.


The biomechanical response of neck under axial compression is defined by Figure 70.


Figure 73. Force-deformation corridor for neck compression when lordosisis removed [Pintar etc., 1995].

16.3 Neck Tensile Loading Requirement

References

Van Ee, C.; et al. 2000. Tensile Properties of the Human Muscular and Ligamentous CervicalSpine. 44th Stapp Car Crash Journal. Paper 2000-01-SC07.

Description

This requirement is used as a secondary requirement for defining human neck response. Eecarried out a series of neck tensile tests using six unembalmed cadaver head-neck complexes at adisplacement rate of 2mm/s. The effects of the cranial end conditions and load eccentricity wereexamined. Corridors were given as force-deformation histories. The corridor developed for thefixed end condition has been selected, since this provides the best setup for establishingrepeatability in testing.

Test Setup

1. The head end is fixed so that no motions is allowed.2. The direction of the loading force is perpendicular to the Frankfurt plane of the head.3. The displacement rate for the loading is 2mm/s;


Figure 74. Force-deformation corridor for neck tensile test under fixedcranial end condition.


The force-deflection response to quasi-static loading in tension of a cadaveric head-neck systemis given by the graph below.

16.4 Neck Torsion Requirement

References

Myers, B.S., McElhaney, J.H., Doherty, B.J., Paver, J.G., Nightingale, R.W., Ladd, T.P. andGray, L. 1989. Responses of the Human Cervical Spine to Torsion, 33rd Strapp Car CrashConference, Paper No. 892437. Wismans, J., and Spenny, C.H. 1983. Performance Requirements for Mechanical Necks inLateral Flexion. Twenty-seventh Stapp Car Crash Conference. SAE Paper # 831613.

Description

This requirement is used as a secondary requirement for defining human neck response. Thetorsion response corridor is based on the study by Myers [1989]. Myers performed a series ofneck torsion tests using unembalmed cadaver cervical spines. High velocity failure tests werecarried out using ramp-to-failure constant velocity displacements. The lower bound (nomuscular action) of the stiffness of the human neck in rotation was derived from these cadavertests. The upper bound was derived by Myers based on analysis of volunteer sled testsperformed by Wismans and Spenny (1983). The target requirement will be based on the


Figure 75. Response corridor of cervical spine undertorsion load

dynamic response, since this corresponds to the environment in which the neck will be tested.

Test Setup

1. The neck assembly is fixed to fixtures at two ends, and is placed horizontally.2. The axial displacement of the neck should be allowed during the torsion.3. The test is performed at a high rotational velocity up to 500 degrees/second.


The torsion response is defined in the graph below.


17. SPINE

References

Cheng, R., Mital, N., Levine, R., and King, A. 1979. Biodynamics of the living human spineduring -Gx impact acceleration. Proc. 23rd Stapp Car Crash Conference, pp 721-763. SAE,Warrendale, Pa.Demetropoulos, C.K., Yang, K.H., Grimm, M.J., Khalil, T.B. and King, A.I., 1998, MechanicalProperties of the Cadaveric and Hybrid III Lumbar Spines, 42nd Stapp Car Crash Conference,Paper No. 983160.Demetropoulos, C.K., Yang, K.H., Grimm, M.J., Artham, K.K. and King, A.I., 1998, High RateMechanical Properties of the Hybrid III and Cadaveric Lumbar Spine in Flexion and Extension,43rd Stapp Car Crash Conference, Paper No. 99SC18.Mital, N., Cheng, R., Levine, R., and King, A. 1978. A new design for a surrogate spine. Proc.7th Experimental Safety Vehicle Conference, pp 427-438. U.S. Government Printing Office,Washington, D.C.Nyquist,, G., and Murton, C.J. 1980. Static bending response of the human lower torso. Proc.19th Stapp Car Crash Conference, pp 513-541. SAE, Warrendale, Pa.Yogannandan, N., Pintar, F, Sances, A. Jr., Maiman, D. and Myklebust, J., Harris, G. and Ray, G.,1988, Biomechanical Investigations of the Human Thoracolumbar Spine, SAE paper No.881331.

Description

At the present time, there are limited biomechanical data which properly define the kinematics ofthe spine under external loading on the whole body. There are test data which measure theresponse of isolated segments of the spine, but it is difficult to extrapolate this information tohow the spine would behave when it is integrated with the complete thorax and pelvis. There arequasi-static data developed by Nyquist and Murton [1975] for the resistance of the whole spinalcolumn from T8 to the pelvis for rotation about the H-point. Dynamic response data wereobtained from tests in a NHTSA study [Cheng, etal, 1979; Mital, etal, 1978] which hadadditional information on the relative motion of the thoracic and lumbar spine segments. Fromthese tests, it was concluded that both segments provided about the same degree of flexibilityand that both of these segments had about the same degree of bending stiffness.

There are also quasi-static data on thoracolumbar spine developed by Yoganandan et al. [1988],and quasi-static data on lumbar spine developed by Demetropoulos [1998]. Demetropoulos et al.also performed some dynamic test on human lumbar spines at a displacement rate 4m/s [1999],but the data were too irregular to be considered useful.


18. THORAX

18.1 Thorax: Regional Coupling under Quasi-static Loading

Reference

Schneider, L., et al. 1992. Design and Development of an Advanced ATD Thorax System forFrontal Crash Environments, Volume 1. UMTRI. DOT Report No. DOT HS 808 138

Description

A measure of the coupling between different regions of the thorax to localized loading has beendeveloped by Cavanaugh [Schneider, 1992]. He statically loaded the chests of three cadaversusing a 5 cm x 10 cm rigid loading plate. Loads were applied at various locations on the sternumand ribcage, and for each loading location, the deflections at eight locations on the anteriorthorax were measured. The locations on the sternum were at the upper, mid, and lower sternum. The locations on the ribs were at the 2nd, 5th, and 8th ribs on both the right and left sides.

Test Setup

1. Subject was placed in a supine position and loading was applied on the denuded ribcage.2. The spine and the posterior ribs were supported bilaterally, approximately 7 cm lateral to

the midline.3. Loading rate ranged from 1.3 mm/s to 100 mm/s and stroke set at 2.5 cm.4. Size of the rectangular loading surface was 5 cm x 10 cm.5. Loading was applied on the upper, mid, and lower sternum, and at costo-chondral

junction of the 2nd, 5th, and 8th ribs. The orientation of the loading surface at the upperand lower ribcage positions resembled the layout of a shoulder belt at these locations.


The deflection responses at the eight thorax locations, to the quasi-static loading are shown inthe tables below. The expected range of relative deflections are given, with the deflection at thepoint of loading being specified as 1.0.

Table 1. Regional deflections to localized thoracic loadingUpper Sternum Load

Right Ribs Sternum Left Ribs

2: 0.6 - 0.8 U: 1.0 2: 0.6 - 0.8

5: 0.3 - 0.5 - 5: 0.3 - 0.5

8: 0.1 - 0.3 L: 0.4 - 0.6 8: 0.1 - 0.3


Mid Sternum Load


2: 0.4 - 0.6 - 2: 0.4 - 0.6

5: 0.4 - 0.6 M: 1.0 5: 0.4 - 0.6

8: 0.3 - 0.5 - 8: 0.3 - 0.5

Lower Sternum Load


2: 0.2 - 0.4 U: 0.2 - 0.4 2: 0.2 - 0.4

5: 0.5 - 0.7 - 5: 0.5 - 0.7

8: 0.4 - 0.6 L: 1.0 8: 0.4 - 0.6

Right 2nd Rib Load


2: 1.0 U: 0.5 - 0.7 2: 0.2 - 0.4

5: 0.2 - 0.4 - 5: 0.0 - 0.2

8: 0.1 - 0.3 L: 0.2 - 0.4 8: 0.0 - 0.2

Right 5th Rib Load


2: 0.1 - 0.3 U: 0.1 - 0.3 2: 0.0 - 0.2

5: 1.0 - 5: 0.1 - 0.3

8: 0.5 - 0.7 L: 0.5 - 0.7 8: 0.1 - 0.3

Right 8th Rib Load


2: 0.0 - 0.2 U: 0.0 - 0.2 2: 0.0 - 0.2

5: 0.4 - 0.6 - 5: 0.1 - 0.3

8: 1.0 L: 0.3 - 0.5 8: 0.1 - 0.3


18.2 Thorax: Regional Loading under Belt Impact

References

Cesari, D.; Bouquet, R. 1990. Behaviour of Human Surrogates Thorax Under Belt Loading. 34th Stapp Car Crash Conference. SAE Paper No. 902310.

Description

Cesari and Bouquet [1990] performed a series of belt impact tests on cadavers to examine thedeflections at different points on the chest under belt loading. A belt with a layout resembling ashoulder belt was pulled dynamically against the thorax and deflection measurements were madeat 11 locations, which included points on the sternum, upper and lower ribcages. Locations bothadjacent to the belt and relatively further away were monitored. The belt was connected to acable and pulley system which could be activated using an impactor.

Test Setup

1. Subjects were supine on a rigid, flat table with belt strap placed diagonally across thechest between the shoulder and the lower ribcage.

2. The two ends of the belt were connected to a cable and pulley system which could be hitby an impactor.

3. Two impactor masses (22.4 kg and 76.1 kg) were used. The impact speeds with thelighter impactor ranged from 2.8 to 9.3 m/s, while impact speeds with the heavierimpactor ranged from 2.5 to 3.5 m/s.

4. Vertical deflections of the ribcage (inside flesh) were taken at the clavicle (opposite belt),upper sternum, lower sternum, anterior portion of the fifth and seventh ribs. Horizontaldeflections were measured on the lateral portion of the eighth rib. Vertical deflectionmeasurements were also taken along the belt path at the left clavicle, mid sternum, andlower ribcage (around the seventh rib).


Preliminary response defined by the belt loading test is relative deflections at different points ofthe thorax. This is given the following table.

Table 2. Regional relative deflections to dynamic belt loadingRight side Sternum Left side

Clavicle: 0.3 - 0.5 Upper: 0.5 - 0.9 Clavicle: 0.5 - 0.9

Rib 5: 0.8 - 1.2 Mid: 1.0 Rib 5: 0.1 - 0.3

Rib 7: 1.0 - 1.4 Lower: 0.8 - 1.1 Rib 7: 0.1 - 0.3

Rib 8 (y): -0.3 - -0.5 -


19. ABDOMEN

19.1 Abdomen: Belt Impact

Reference

Hardy, W.; Schneider, L. 2001. Development and Refinement of Abdominal-ResponseCorridors. UMTRI Report UMTRI-2000-10.

Rouhana, S.W., Viano, D.C., Jedrzejczak, E.A., and McCleary, J.D., 1989. AssessingSubmarining and Abdominal Injury Risk in the Hybrid III Family of Dummies. 33rd Stapp CarCrash Conference, Paper No.892440.

Description

Hardy and Schneider [2001] have recently developed a corridor for force vs. penetration of abelt into the abdomen at the level of the umbilicus. Tests were conducted using three cadaverswith repeat tests on each. This corridor is different from that developed by Rouhana [1989], whohad scaled the data from porcine tests. In the new tests, the belt essentially circled half theabdomen while in the previous tests the belt was straight and was only in tangential contact atthe beginning of the load sequence. It is felt that the new configuration is closer to the setupseen by an occupant in a vehicle.

Test Setup

1. Subjects were placed in a seated, upright position with legs positioned forward and thehands placed in front of and above the head. The subject was placed on a curved,Teflonskid.

2. A standard seat belt (6% stretch) was placed about the mid-abdominal region,approximately at the level of the umbilicus. The webbing was in contact with theanterior aspect of the abdomen and routed around the sides so that the two ends weretangent to the lateral aspect of the abdomen.

3. The belt was loaded on to the abdomen with a peak loading rate of about 3 m/s with theloading pulse having the shape of a half sine wave.

4. Measurements were made of the loads at the two ends of the seat belt and the penetrationinto the abdomen.


The response to belt loading about the mid abdomen is given by the following


Figure 76. Corridors for belt impact response with mid abdomen at 3 m/s.

19.2 Abdomen: Airbag Impact

Reference

Hardy, W.; Schneider, L. 2001. Development and Refinement of Abdominal-ResponseCorridors. UMTRI Report UMTRI-2000-10.

Description

Hardy and Schneider [2001] have recently developed a corridor for force vs. penetration of anairbag against the abdomen. They developed a surrogate device for the airbag, which was a thin-walled, aluminum cylinder with the cylindrical face impacting the abdomen. The ends of thecylinder were capped with rounded faces to eliminate sharp loads as the cylinder penetrated theabdomen. The fixture is meant to simulate the early stage of a deploying airbag. The device wasdriven by a pneumatic cylinder and excess pressure is allowed to be vented to the atmosphere. The impact speed was set at about 13 m/s. The final corridor was based on the results from testson three cadavers.

Test Setup

1. Subjects were placed in a seated, upright position with legs positioned in front of thecadaver, and the hands placed in front and above the head. The backs of the subjectswere supported.

2. The center of the surrogate airbag device was aligned with the umbilicus at the level ofthe L3 verterbra.

3. The impactor face was the cylindrical side of 7.6 cm diameter tube of length 20 cm, andmass 1 kg. The impactor face was initially placed in front of the abdominal surface.


Figure 77. Force-penetration corridors for airbag impact with abdomen at13 m/s.

4. Impact speed was set at 13 m/s.5. Measurement of the impact load due to the surrogate device surface was made using the

acceleration of the device and the internal pressure within the cylinder.


The response to airbag impact to the mid abdomen is given by the following.


20. FACE

20.1 Cylindrical Impactor Response

References

Bermond, F., Kallieris, D., Mattern, R., Ramet, M., Bouquet, R., Caire, Y. and Voiglio, É., 1999,Human Face Response at an Angle to the Fore-aft Vertical Plane Impact, IRCOBI Conference1999, pp.121-132.

Description

The corridors here are based on the test data by INRETS [Bermond, 1999]. A total of 57 facialimpact tests were performed using a cylindrical impactor simulating a steering wheel rim. Theimpacts were delivered to the forehead, zygoma and maxilla at an angle of 30 degrees to thefore-aft vertical (sagittal) plane in order to represent typical accidents conditions. The corridorshere are impactor peak force and resultant acceleration at head C.G. vs impact velocity for threedifferent regions of the face.

Test Setup

The test configuration of this test is defined as:1. The impactor is guided, the mass of the impactor is 17.0kg.2. A 2.25 cm diameter steel cylinder is mounted horizontally at the far extremity of the

impactor and hits the face directly.3. The impact speed is at 3m/s.4. The impact should be delivered at an angle of 30 degrees with regard to the fore-aft

vertical plane.


The corridors for forehead, zygoma and maxilla are shown in Figures 75 - 80.


Figure 78. Corridor for impactor peak force whenhitting zygoma

Figure 79. Corridor for resultant peak acceleration athead CG when hitting zygoma

Figure 80. Corridor for impactor peak force whenhitting forehead

Figure 81. Corridor for resultant peak acceleration athead CG when hitting forehead

Figure 82. Corridor for impact peak force whenhitting maxilla

Figure 83. Corridor for resultant peak acceleration athead CG when hitting maxilla



21.1 Dynamic Inversion/Eversion

References

Jaffredo, A., et al. 2000. Cadaver Lower Limb Dynamic Response in Inversion-Eversion. 2000International IRCOBI Conference on the Biomechanics of Impact. pp 183-194.

Description

Jaffredo, et al. [2000] studied the dynamic response of the ankle in inversion-eversion. Six pairsof cadaver lower limbs were tested with six tests in inversion and six tests in eversion. Accelerations, angular velocities, forces, and moments were measured on the forefoot, heel, andleg. Force balance was computed at the subtalar joint. The contribution of the fibula was alsomeasured and the angle between the forefoot and heel computed. The subtalar joint moment wasevaluated with respect to the xversion angle of the forefoot and of the heel.

In these tests, the lower extremity was amputated at the knee an mounted to the test apparatus. All musculature was preserved. In half the tests, only the forefoot was fixed to the plate, and inthe other half, both the forefoot and heel were fixed.. The fixation points corresponded with theanatomic support points of the foot defined by previous X-rays.

Load cells were planted within the tibia and the fibula to determine their contributions to thetotal axial and shear forces and to the X-axis moment. They also found that the fibula couldcontribute up to 17.5% of the total moment in the subtalar joint and 50% of the total axial force. This indicated that the fibula can play a significant role in the overall response of the ankle. Inseveral previous studies, the contribution of the fibula was ignored, since only the tibia load cellwas inserted. They also estimated the axial force due to passive muscle in inversion andeversion. It was linear as a function of xversion angle. A resistive force of 150 N was generatedfor a calcaneal angle of 15°. Unlike dorsiflexion, the passive musculature had no significantcontribution to the xversion moment.

Test Setup

1. Lower extremity amputated at knee. Proximal end fixed to a slider, while foot fixed to arigid, rotating plate, which allows foot to be moved in inversion/eversion.

2. Foot positioned so plate rotation axis is superimposed on the xversion axis defined fromX-rays.

3. Foot set at about 90 deg relative to tibia.4. Impact ball of foot at 2 m/s.5. Measure forces and moments at distal tibia and fibula, and angle of rotation of calcaneous

about center of xversion. Compute moment-angle response at the subtalar joint.


Figure 84. Moment vs angle response for dynamicinversion (and comparison to static).

Figure 85. Moment vs angle response for dynamiceversion (and comparison with static).


The moment-angle response corridors for dynamic inversion and eversion, estimated from thecadaver tests are shown in the following graphs. The angle is the measured angle of thecalcaneous relative to the tibia. Comparison with the previous quasi-static response curves areshown.

BIOMECHANICAL RESPONSE REQUIREMENTS OF THE THOR … · THE THOR NHTSA ADVANCED FRONTAL DUMMY...

Documents

Transcript of BIOMECHANICAL RESPONSE REQUIREMENTS OF THE THOR … · THE THOR NHTSA ADVANCED FRONTAL DUMMY...